The Home Run Pain Scale

Analyzing how homer timing inflates and deflates pitcher ERAs

Colin’s Mens Rea

Colin Rea's 2025 season was the kind that keeps a veteran in a rotation without anyone thinking twice about it. He threw 159.1 innings with a 3.95 ERA and allowed 20 home runs, a steady back-of-the-rotation line. In 2026 his ERA rose to 4.34 over 137 innings, his strikeout rate slipped from 19.2% to 17.2%, and his on-base percentage allowed ticked up from .309 to .322. He allowed 21 homers, one more than the year before. On the surface it reads as a modest slip, the kind usually pinned on a few lost strikeouts, and that explanation is only part of the story.

What the ordinary stats don't show is where the extra runs came from. Rea's homers cost him 31 runs in 2025 and 37 in 2026, even though he allowed nearly the same number of them. In 2025 he gave up 9 solo homers, 11 two-run homers, and not a single three-run homer or grand slam. In 2026 he gave up 8 solo homers, 10 two-run homers, and 3 three-run homers. At the league rate of about 1.58 runs per homer, his 2025 homers scored almost exactly what they should have, while his 2026 homers cost him about four runs more than average, and those three swings with two men aboard are where the extra runs came from.

One way to see through that is a homer-neutral ERA, which keeps every run a pitcher allowed except the ones his homers scored and replaces those with his homer count times the league-average runs per homer (1.57 in 2025, 1.58 in 2026). By that measure Rea went from 3.98 to 4.08, essentially the same pitcher in both seasons. Of the 0.39 he added to his ERA, roughly 0.29 came down to who was standing on base when the ball left the park, and only about a tenth of a run reflects him pitching worse. That raises a simple question with a less obvious answer: how often does a homer actually come with runners aboard, and how much does that luck move a pitcher's season?

Solo Shots vs 3-Run Johnsons

Most home runs are solo shots, and it isn't close. Of every homer hit in the 2026 regular season, 58.0% came with the bases empty, 28.3% were two-run homers, 11.4% were three-run homers, and just 2.3% were grand slams. That works out to 1.58 runs per home run across the league, and even when a homer does come with men on, it carries an average of 1.38 runners with it. The three-run homer and the grand slam feel like the defining home runs of a season because they are the ones that flip games, but together they make up only about one homer in every seven. That rarity is exactly what makes them so lumpy. A pitcher who allows 20 homers might see two or three of them come with two men on, or he might see six, and the difference can be a dozen runs on his ledger.

Nobody in lived in pain in 2026 like Mike Burrows did, who allowed seven three-run homers on only 21 total home runs. The pitchers tied behind him needed as many as 35 homers to reach six.

Pitcher

HR

Solo

2-run

3-run

Slams

Runs on HR

Runs per HR

Mike Burrows

21

9

5

7

0

40

1.90

Jameson Taillon

27

17

4

4

2

45

1.67

Jeffrey Springs

35

17

12

6

0

59

1.69

Shota Imanaga

35

25

4

6

0

51

1.46

Jack Perkins

23

10

7

6

0

42

1.83

Kevin Gausman

26

13

8

4

1

45

1.73

Spencer Arrighetti

18

7

6

4

1

35

1.94

Nolan McLean

17

9

3

4

1

31

1.82

Bryan Abreu

9

3

1

4

1

21

2.33

Miles Mikolas

27

15

7

5

0

44

1.63

The list has a few wrinkles worth pointing out. Shota Imanaga is tied for third in three-run homers and slams, yet his runs per homer finished below the league average because 25 of his 35 homers were solo shots, so a pile of crooked numbers doesn't always add up to an expensive homer season. Bryan Abreu allowed only nine homers out of the bullpen, and five of them came with at least two men on. Gabriel Hughes, just off this list, allowed seven homers all year and two were grand slams. Among pitchers who allowed at least 15 homers, the range runs from Payton Tolle at 1.13 runs per homer to Spencer Arrighetti at 1.94, a gap that over Arrighetti's 18 homers is worth about 15 runs.

The other end of the scale belongs largely to the game's best pitchers. Sonny Gray allowed 16 homers and 14 of them were solo shots. Zack Wheeler and Paul Skenes each saw at least 80% of their homers come with the bases empty. That is no accident, since elite pitchers rarely let anyone on base, and a pitcher who keeps the bases clear will naturally give up more of his homers with nobody aboard. Limiting homers to solo shots is part of what being great looks like. The question is whether anything beyond that carries over from one season to the next.

Does This Mean Anything (No)

As expected very little of this HR noise carries over season-over-season: among pitchers who allowed at least 15 homers in both 2025 and 2026, runs per homer had a year-to-year correlation of just 0.12, where 1 would mean one season perfectly predicts the next and 0 would mean it predicts nothing at all. Compare that with the things pitchers clearly do control: home run rate carried over at 0.43 and on-base percentage allowed at 0.29. Pitchers have real say over how many homers they allow and how many runners they put on, but almost none over whether a particular homer arrives with the bases empty or loaded. In practical terms, a correlation of 0.12 means you should expect close to 90% of a pitcher's runs-per-homer luck to disappear the following year.

Not every bad season washes out that cleanly. Jameson Taillon went from a 3.68 ERA over 129.2 innings in 2025 to a 5.65 ERA over 86 innings in 2026, split between the Cubs and Blue Jays. His 24 homers in 2025 included no three-run shots or grand slams and cost him just 1.33 runs apiece, while six of his 27 homers in 2026 came with two or three men aboard and cost him 1.67 apiece. His homer-neutral ERA goes from about 4.08 to 5.41, so timing explains roughly two-thirds of a run of a nearly two-run collapse. The rest was real: he allowed a homer to 7.3% of the batters he faced, up from 4.6%, and his on-base percentage allowed jumped from .263 to .321. Homer luck flattered his 2025 and piled onto his 2026, but it didn't create the decline.

Stripping out that one source of noise makes ERA a bit more stable from year to year. Among pitchers who threw at least 80 innings in both seasons, ERA carried over at 0.26 and homer-neutral ERA at 0.30. K-BB% (strikeout rate minus walk rate), which ignores sequencing entirely, carried over at 0.58. Homer timing is one identifiable reason ERA bounces around from season to season, and the rest of that gap is the same kind of sequencing on everything else, singles and doubles that land with runners on or don't, plus the defense behind the pitcher (you find more on this from my Razzball article here). It is also why FIP (Fielding Independent Pitching, which builds an ERA-scale number from strikeouts, walks, hit batters, and homers) counts every home run the same whether it was a solo shot or a grand slam. A pitcher's ERA is telling you what happened, but how many runners were on base when the ball left the yard is mostly a coin flip, and the steadier numbers are the ones we use time after time.

Get Balls & Sticks in your inbox

Subscribe for free to get new posts by email as soon as they publish.